Electrochemical regeneration of NADH enhanced by platinum nanoparticles

Article abstract of DOI:10.1002/anie.200703632

(Figure Presented) Wireless communication: Platinum nanoparticles (nPt) in an electrolyte enhance electron transfer from the electrode to NAD⁺ during the indirect electrochemical regeneration of NADH (see picture). The intermediate nPt-H_ads, formed at negative potential, helps the turnover of the primary mediator M by donating a proton and an electron in a kinetically favorable way.

Full text of DOI:10.1002/anie.200703632

Angewandte
Chemie
DOI: 10.1002/anie.200703632
Nanoparticles
Electrochemical Regeneration of NADH Enhanced by Platinum
Nanoparticles**
Hyun-Kon Song, Sahng Ha Lee, Keehoon Won, Je Hyeong Park, Joa Kyum Kim, Hyuk Lee,
Sang-Jin Moon, Do Kyung Kim, and Chan Beum Park*
Herein, we report the application of nanoparticulate platinum
(nPt) to enhancing the heterogeneous electron transfer
between NAD+ (nicotinamide adenine dinucleotide, oxidized
form) and electrodes in the presence of an organometallic
The rhodium complex M was successfully used as an
electron shuttle for NAD+ in electrolyte, which improved the
kinetics of NADH regeneration.[7–9] The active reduced form
Mred2 that enables NADH to be generated is made through a
mediator.
(Pentamethylcyclopentadienyl-2,2’-bipyridine-
typical electrochemical/chemical (EC) process (Scheme 1).
chloro)rhodium(III) (M = [Cp*Rh(bpy)Cl]+; Cp* = C5Me5,
bpy = 2,2’-bipyridine) was used as a primary mediator to
shuttle electrons between NAD+ and electrodes. nPt func-
tioned as a homogeneous catalyst and also as a secondary
mediator to improve the turnover kinetics of M.
Pyridine nucleotides (NAD(P)H) or their oxidized coun-
terparts (NAD(P)+) are used as cofactors that are critically
required for redox reactions catalyzed by various oxidore-
ductases.[1,2] In biocatalytic reactions, NADH should be
regenerated to allow the enzymes to continue their turnover.
Electrochemical regeneration has been chosen as an attrac-
tive strategy that is an alternative to enzymatic regenera-
tion.[3] In electrochemical regeneration, however, the first
drawback to overcome is the slow electron transfer between
NAD+ and the electrodes, even at a potential where the
reduction of NAD+ into NADH is thermodynamically
favorable. The use of a homogeneous mediator to shuttle
electrons between electrodes and NAD+ can be one solution
to solve the problem.[4–6]
Scheme 1. Molecular structures of three different electrochemical
states of M.
Mox is reduced to Mred1 by accepting two electrons from the
electrodes (E step). Successively, Mred1 is chemically con-
verted into Mred2 without any change in the total number of
electrons, by taking up one proton from solution (C step).
NADH is generated from NAD+ with the active form Mred2 by
accepting one proton plus two electrons from Mred2 and
returning Mred2 to the initial state Mox (Scheme 2).
The chemical step from Mred1 to Mred2 (the uptake of a
proton into the ligand sphere of Mred1) is the rate-determining
step of NADH generation, specifically at high rates, even
though it was reported to proceed quite fast.[8] Figure 1a
shows cyclic voltammograms obtained at various scan rates.
The intensity of the anodic peak increased with scan rate,
whereas the peak was not apparently observed at scan rates
[*] S. H. Lee, J. H. Park, Prof. D. K. Kim, Prof. C. B. Park
Department of Materials Science and Engineering
Korea Advanced Institute of Science and Technology
373-1 Guseong-dong, Yuseong-gu, Daejeon 305-701 (Korea)
Fax: (+82)42-869-3310
E-mail: parkcb@kaist.ac.kr
Prof. K. Won
Department of Chemical and Biochemical Engineering
Dongguk University
26 Pil-dong 3-ga, Jung-gu, Seoul 100-715 (Korea)
J. K. Kim, Dr. H. Lee, Dr. S.-J. Moon
Korea Research Institute of Chemical Technology (KRICT)
100 Jang-dong, Yuseong-gu, Daejeon 305-343 (Korea)
Dr. H.-K. Song[+]
Division of Engineering
Brown University, Providence, RI 02912 (USA)
[+] Present address: Battery R&D, LG Chem Ltd.
Research Park, 104-1 Moonji-dong, Yuseong-gu, Daejeon 305-380
(Korea)
[**] This work was supported by the Korea Energy Management
Corporation (2005-C-CD11-P-04) and the Korea Research Founda-
tion (KRF-2006-331-D00113).
Scheme 2. Indirect electrochemical regeneration of NADHwith the
primary mediator M and its enhancement by proton and electron
transfer from nPt to M. Dashed arrows indicate electron transfer.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2008, 47, 1749 –1752
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1749

Communications
mediator strategy that includes nPt as well as M. These two
mediators are reduced to nPt-Hads and Mred1 at ꢀ0.8 V. nPt-
Hads returns to nPt by donating a proton to Mred1 and an
electron to Mox or NAD+.
Figure 2a and b shows the change of voltammetric
features before and after the addition of nPt to the single-
mediator systems (M or M + NAD+). The reduction potential
Figure 1. a) Cyclic voltammograms of M (500 mm) in phosphate buffer
(100 mm) at pH8.2. Scan rates are indicated in mVs ꢀ1. b) Scan rate
dependency of cathodic and anodic peak currents obtained from (a).
Slopes of the lines were estimated at 0.5.
less than 100 mVsꢀ1. At slow anodic potential sweep, the
C step proceeds faster than the E step so that there is no
chance for Mred1 to donate electrons to the electrode. In fast
scans (rate above 500 mVsꢀ1), however, Mred1 is electro-
chemically oxidized to Mox before the C step proceeds, as the
electron-transfer rate between Mred1 and electrodes is con-
trolled by the scan rate and is faster than the rate of the
chemical reaction.
The anodic peak current ip followed the dependency on
scan rate of conventional faradaic processes (ip / [scan
rate]0.5) only at scan rates higher than 500 mVsꢀ1, which
indicates that Mred1 was totally converted to Mox by the E step
(Figure 1b). The peak currents at scan rates less than
500 mVsꢀ1 deviated from the extrapolated lines fitting peak
currents at the three largest scan rates, because Mred1 was
partly converted to Mred2 by the C step. Therefore, the kinetics
of proton uptake in the C step should be enhanced to achieve
efficient formation of Mred2, which is active for NADH
generation.
Platinum has been extensively used to reduce protons in
electrolytes to hydrogen (hydrogen evolution reaction, HER)
and also to oxidize hydrogen to protons in fuel cells.[10,11] The
main reason that makes the platinum catalyst superior to
other alternative metals is that protons are adsorbed onto
platinum atoms. The intermediate state Pt-Hads makes the H+/
H2 reaction kinetically more favorable, which results in a
decrease of overpotential.
Metal–Hads species, including Pt-Hads, were reported to be
able to function as a reducing agent for organic molecules,
markedly in their nanoparticulate form. Platinum nanopar-
ticles with adsorbed hydrogen atoms (nPt-Hads) were used to
reduce the lucigenin cation to its monocation radical in the
potential range of the HER.[12] Also, other metal nano-
particle–Hads species were reported to work as a reducing
agent:[13,14] nAg-Hads to reduce CH2Cl2 to CH3Cl or Tl+ to Tl0;
nPd-Hads to reduce Pt4+ or Pt2+ into Pt0.
Based on an understanding of the intermediate Pt-Hads, we
added nPt to the single-mediator strategy of M + NAD+. nPt
was introduced to play two functional roles in our tandem
strategy: 1) the homogeneous catalyst responsible for cata-
lyzing the proton uptake reaction of Mred1 to Mred2, and 2) the
secondary mediator to shuttle electrons from electrodes to
Mox. Scheme 2 shows the working mechanism of our tandem-
Figure 2. a,b) Cyclic voltammograms of solutions of a) M and
b) nPt+M in the absence and presence of NAD+. nPt (0.6 mm), M
(0.5 mm), and NAD+ (0.5 mm) were used in phosphate buffer
(100 mm) at pH7.0. GC electrodes were used as working electrodes.
The potential was scanned at 100 mVsꢀ1. Inset in (b): transmission
electron microscopy image of nPt. c,d) Conversion rates at ꢀ0.8 V of
c) NAD+ to NADHin the absence and presence of nPt and d) nPt to
nPt-Hads in the absence and presence of NAD+. The rates (=Di/nFA;
n=number of electrons, F=Faradaic constant, A=electrode area)
were calculated from the difference of the cathodic currents at ꢀ0.8 V
(in a,b) between M+NAD+ and M for “ꢀnPt” in (c); between
nPt+M+NAD+ and nPt+M for “+nPt” in (c); between nPt+M and
M for “ꢀNAD+” in (d); and between nPt+M+NAD+ and M+NAD+
for “+NAD+” in (d).
at the cathodic peak current (Epc) of M was estimated at
ꢀ0.7 V in the absence or presence of NAD+ (Figure 2a).
After addition of nPt, the cyclic voltammograms were totally
changed. The cathodic and anodic waves shown in Figure 2b
arise mostly from adsorption and desorption of protons on
nPt with Epc = ꢀ0.85 to ꢀ0.9 V and the oxidation potential at
the anodic peak current Epa = ꢀ0.52 V.[10]
The conversion rates of NAD+ to NADH (Figure 2c) or
nPt to nPt-Hads (Figure 2d) on electrodes were calculated
from the difference of cathodic currents at ꢀ0.8 V (the
working potential used to generate NADH). The addition of
nPt enhanced the rate of NADH generation 25 times (rate
difference = 3.82 nmolsꢀ1 cmꢀ2). Also, the rate of nPt reduc-
tion on electrodes increased 25% in the presence of NAD+
when compared with that in the absence of NAD+ (rate
difference = 7.64 nmolsꢀ1 cmꢀ2).
The amount of NADH was measured spectroscopically at
ꢀ0.8 V in the absence or presence of various concentrations
of nPt. The amount increased with nPt concentration and
1750
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 1749 –1752

Angewandte
Chemie
nanoparticulate feature. The possibility that the increase in
the amount of generated NADH is mainly a result of the
increase in surface area through adsorption of nPt on the
working electrodes should be rejected, because there were no
significant differences in the amounts of NADH when using
bare and nPt-adsorbed GC electrodes.
Figure 3. Temporal change of NADHconcentration ( CNADH) in the
absence or presence of nPt. NAD+ (1 mm) and M (500 mm) were
used; the concentrations of nPt are indicated. The solutions for all
*
experiments were stirred at 340 rpm (except for , when the mixture
was not stirred). For electrodes and buffer solutions see Figure 2.
stirring rate (Figure 3). The concentration of NADH mea-
sured 5 h after the potential was applied increased five times
in the presence of 1.2 mm nPt when compared with that in the
absence of nPt. The NADH generated with nPt and M proved
to work effectively with glutamate dehydrogenase, an oxidor-
eductase, for the synthesis of l-glutamate from a-ketogluta-
rate in preliminary experiments.
To confirm the nanoparticulate contribution of nPt to the
generation of NADH, various flat and bulky disk electrodes
including platinum were tested as the working electrode for
NADH generation in the presence of M. Avery small amount
of NADH was generated on these bulky disk electrodes
(Figure 4a). For comparison, the amount of NADH gener-
Figure 5. Concentration of NADHgenerated on a GC working elec-
trode in a solution including M, nPt, or nPt+M. The solution was
stirred at 340 rpm at ꢀ0.8 V for 1 h prior to addition of NAD+ (1 mm).
A potential was not applied after adding NAD+. The asterisk indicates
that the same amount of NADHwas produced when M and NAD+
were added. nPt (0.6 mm) and M (0.5 mm) were used in phosphate
buffer (100 mm) at pH7.0.
To confirm the roles of nPt in the proposed mechanism,
prereduced solutions were used to generate NADH without
applied potential (Figure 5). If Mox in solution were totally
reduced to Mred2, 0.5 mm NADH would be generated after
addition of NAD+ to the prereduced solution. However, the
amount of NADH generated in the solution of M was
estimated as a very dilute concentration of less than 0.01 mm.
nPt directly reduced NAD+ to NADH even if a very small
amount of NADH was generated. There were no significant
differences of the NADH concentration between solutions
including only M and only nPt. On the other hand, a
synergistic effect was observed in the mixture of nPt and M.
The amount of NADH generated increased about 20 times
(0.125 mm NADH), thus indicating that nPt enhanced the
formation of Mred2
.
Figure 4. a) Concentration of NADHgenerated on various working
electrodes in a solution of M and NAD+, stirred at 340 rpm, at ꢀ0.8 V
for 2 h. b) Temporal change of concentration of NADHgenerated on
Pt working electrodes in the absence or presence of nPt. Solutions
were stirred at 340 rpm at ꢀ0.8 V during NADHgeneration. M
(0.5 mm) and NAD+ (1 mm) were used in phosphate buffer (100 mm)
at pH7.0 for all experiments.
In conclusion, nPt was used as a homogeneous catalyst
and simultaneously as a secondary mediator for NADH
regeneration in the presence of the primary mediator M. It
enhanced the rate of NADH generation by donating protons
and electrons to M. We expect that the use of nPt could be
extended to the reduction of other chemicals, even in proton-
deficient environments (high pH).
ated on glassy carbon (GC) disk electrodes in the presence of
1.2 mm nPt under the same conditions was 0.2 mm (Figure 3).
The platinum disk electrode was even less efficient for
NADH generation than GC and gold disk electrodes. On the
other hand, the addition of nPt to the solution including M
and NAD+ resulted in an increase of the amount of NADH
generated on the platinum disk electrode (Figure 4b). There-
fore, the catalytic (proton donation) and reducing (electron
donation) power of nPt can be said to originate from its
Experimental Section
The rhodium complex M was synthesized by the method of Kolle and
Gratzel.[15] The nPt was prepared by citrate reduction of potassium
tetrachloroplatinate.[16,17] An aqueous solution of sodium citrate
(30 mL, 680 mm) was added to a boiling aqueous mixture of K2PtCl4
(120 mL, 11.5 mm) and polyvinylpyrrolidone (3 g, molecular weight
10k) for 4 h.
Angew. Chem. Int. Ed. 2008, 47, 1749 –1752
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1751

Communications
To sweep the potential in cyclic voltammograms or to apply
[6] J. Wang, P. V. A. Pamidi, M. Jiang, Anal.Chim.Acta 1998, 360,
constant potential for generating NADH, a single-compartment cell
was configured with three electrodes: a GC disk (working, 0.03 cm2),
a platinum wire (counter), and an Ag/AgCl (reference, 0.197 V versus
normal hydrogen electrode) connected to a potentiostat/galvanostat
(EG&G, Model 273A). All potentials are reported versus Ag/AgCl.
The concentration of NADH was estimated from the difference of the
absorbance at 340 nm before and after adding NAD+.
171.
[7] R. Ruppert, S. Herrmann, E. Steckhan, Tetrahedron Lett. 1987,
28, 6583.
[8] F. Hollmann, A. Schmid, E. Steckhan, Angew.Chem. 2001, 113,
190; Angew.Chem.Int.Ed. 2001, 40, 169.
[9] K. Vuorilehto, S. Lütz, C. Wandrey, Bioelectrochemistry 2004, 65,
1.
[10] N. Nanbu, F. Kitamura, T. Ohsaka, K. Tokda, J.Electroanal.
Chem. 2000, 485, 128.
Received: August 9, 2007
[11] A. Henglein, B. Lindig, J. Westerhausen, J.Phys.Chem. 1981, 85,
1627.
Revised: October 10, 2007
Published online: January 25, 2008
[12] J. Z. Guo, H. Cui, S. L. Xu, Y. P. Dong, J.Phys.Chem.C 2007,
111, 606.
[13] A. Henglein, J.Phys.Chem. 1979, 83, 2858.
[14] Y. Wang, N. Toshima, J.Phys.Chem.B 1997, 101, 5301.
[15] F. Hollmann, B. Witholt, A. Schmid, J.Mol.Catal.B 2002, 19–20,
167.
[16] D. N. Furlong, A. Launikonis, W. H. F. Sasse, Chem.Soc.Fara-
day Trans.1 1984, 80, 571.
[17] I. Okura in Encyclopedia of Nanoscience and Nanotechnology,
Vol.8 (Ed.: H. S. Nalwa), American Scientific Publishers, 2004,
p. 41.
Keywords: electron transfer · homogeneous catalysis · NADH·
nanoparticles · platinum
.
[1] E. Siu, K. Won, C. B. Park, Biotechnol.Prog. 2007, 23, 293.
[2] W. A. van der Donk, H. Zhao, Curr.Opin.Biotechnol. 2003, 14,
421.
[3] F. Hollmann, A. Schmid, Biocatal.Biotransform. 2004, 22, 63.
[4] H. Jaegfeldt, A. Torstemsson, L. Gorton, G. Johansson, Anal.
Chem. 1981, 53, 1979.
[5] J. Wang, J. Liu, Anal.Chim.Acta 1993, 284, 385.
1752
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 1749 –1752